Introduction

Last week, Wikibon and SiliconAngle.tv covered EMC’s release of the company’s new VSPEX proven infrastructure product. VSPEX provides prebuilt, pretested, and prevalidated units of converged infrastructure targeted at the SMB and midmarket space. However, where VSPEX starts to differ from some of the other converged infrastructure plays that are out there is in its flexibility. With VCE's Vblock, for example, there is a standard set of hardware, with customization limited to capacity tweaking. Customers must use the underlying vendors, for example, and even the ability to scale up or scale down is limited.

History

Before I go much deeper into a discussion on converged infrastructure, let me back up a bit and talk about a couple of problems that have plagued IT departments for a long time, particularly those in the SMB space. Moreover, these departments may not even know that these have been problems.

These are islands, sprawl, and inconsistency. Even though many SMBs have embraced virtualization, others have not. Believe it or not, I hear stories about IT directors who believe that virtualization is a flash in the pan and nothing more than a fad that will pass with time. I’ve seen these environments up close, and they demonstrate many of the problems that have been solved by virtualization. I’ve seen racks of single-purpose servers--physical sprawl, all with local storage--islands of data, and all managed individually. While this model is functional, these organizations are losing out on the efficiencies and opportunities that come with virtualization.

From a consistency perspective, these environments can be well-managed, but they provide more opportunity for introduction of inconsistency. Further, each system is often managed individually, creating administrative overhead. This translates into additional time spent managing the infrastructure. From a long-term perspective, that’s not a good way to manage valuable IT resources.

Virtual environments have their own challenges. Speaking from experience, SMBs usually start by dipping their toes in the virtualization waters through a server consolidation project and, once they see the benefits, expand these projects to encompass more services. However, SMBs still need to put together infrastructures that can support these virtual environments. This often means choosing a hypervisor, buying servers, then looking for shared storage that can support the required workloads and allow the organization to leverage virtualization’s benefits.

Convergence/proven is the new black

Side note: EMC differentiates between the terms “converged” and “proven” as it relates to infrastructure. For the purposes of this article, we’re going to consider them synonyms. However, be aware that “converged” in EMC’s sense means something like Vblock, which is architected by the VCE coalition, and “proven” indicates an architecture that has ben validated by EMC.

As IT organizations make moves to streamline overall operations in order to better focus on business-facing initiatives, infrastructure is an area ripe for improvement. Infrastructure requires ongoing attention, management, correction, replacement, and upgrade. It takes teams of people to handle and, while infrastructure is the lifeblood of the business, the business wants to keep these overhead costs low in favor of more bottom line-driven needs.

Enter converged infrastructure. Converged infrastructure provides organizations with the opportunity to acquire units of infrastructure on an as-needed basis that can be managed centrally. Better yet, in most cases these converged infrastructure entities are sold as a single product and supported by a single vendor from the bottom of the stack to the top. Are you having a problem with the storage? Are you having a problem with the hypervisor? The solution to both problems is but one phone call away.

Convergence itself is changing

To be fair, this idea of infrastructure convergence is not a new one, even for EMC. As I mentioned before, Vblocks have now been around for a couple of years, and other vendors have jumped on this emerging infrastructure bandwagon.

However, where Vblock was pretty strict in its configuration, VSPEX is not. VSPEX provides customers with significant choice, making it more accessible to more companies. EMC has created a good partner ecosystem, which includes, obviously, VMware and Cisco, but adds Microsoft, Dell, HP, IBM, Citrix, and Brocade. It should also be obvious that the only part of the solution that you can’t customize is the storage layer, although you do have choice in the EMC portfolio.

Beyond EMC

Although EMC has been a pioneer in the converged infrastructure space and VSPEX provides impressive flexibility, many vendors are proving converged infrastructure opportunities, including:

• The Virtual Computing Environment Company (VCE) VCE provides the VBlock Series 300 and Series 700 platforms. Vblock is built on EMC (VNX, VMAX), Cisco (UCS, Nexus, MDS) and VMware hardware and software.
• Dell For SMBs, Dell’s vStart is a good starting point. vStart is available in three flavors, the models 50, 100 and 200. As you might expect, Dell’s vStart uses Dell hardware, including PowerEdge servers, EqualLogic storage arrays, and PowerConnect switches. Dell works with multiple hypervisor vendors.
• NetApp NetApp’s converged play is called a FlexPod. With NetApp storage at the core, NetApp has partnered with Cisco and VMware to provide the compute, networking, and software parts of the stack.
• IBM IBM recently announced its PureSystems initiative, which helps IT organizations focus on project delivery rather than infrastructure.
• HP HP’s converged infrastructure (POD) leverages HP’s line of hardware and third-party virtualization software to streamline IT infrastructure operations.

Action Item: With converged infrastructure solutions that run the gamut from very small to very large, CIOs that want to make it easier to manage their environments would do well to give converged infrastructure solutions a once over. I see these kinds of solutions as a part of the broader move toward IT departments becoming service brokers. The less time spent on managing “keeping the lights on” the more time that can be spent on business-critical activities.

Footnotes: